No Cover Image

Conference Paper/Proceeding/Abstract 1190 views 368 downloads

Learning Discriminatory Deep Clustering Models

A. Alqahtani, Xianghua Xie Orcid Logo, J. Deng, Mark Jones Orcid Logo

Computer Analysis of Images and Patterns, Volume: 11678, Pages: 224 - 233

Swansea University Authors: Xianghua Xie Orcid Logo, Mark Jones Orcid Logo

Check full text

DOI (Published version): 10.1007/978-3-030-29888-3_18

Abstract

Deep convolutional auto-encoder (DCAE) allows to obtain useful features via its internal layer and provide an abstracted latent representation, which has been exploited for clustering analysis. DCAE allows a deep clustering method to extract similar patterns in lowerdimensional representation and fi...

Full description

Published in: Computer Analysis of Images and Patterns
ISBN: 978-3-030-29887-6 978-3-030-29888-3
ISSN: 0302-9743 1611-3349
Published: 2019
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa50907
first_indexed 2019-06-24T14:56:20Z
last_indexed 2020-08-19T03:13:36Z
id cronfa50907
recordtype SURis
fullrecord <?xml version="1.0"?><rfc1807><datestamp>2020-08-18T12:40:15.7748643</datestamp><bib-version>v2</bib-version><id>50907</id><entry>2019-06-24</entry><title>Learning Discriminatory Deep Clustering Models</title><swanseaauthors><author><sid>b334d40963c7a2f435f06d2c26c74e11</sid><ORCID>0000-0002-2701-8660</ORCID><firstname>Xianghua</firstname><surname>Xie</surname><name>Xianghua Xie</name><active>true</active><ethesisStudent>false</ethesisStudent></author><author><sid>2e1030b6e14fc9debd5d5ae7cc335562</sid><ORCID>0000-0001-8991-1190</ORCID><firstname>Mark</firstname><surname>Jones</surname><name>Mark Jones</name><active>true</active><ethesisStudent>false</ethesisStudent></author></swanseaauthors><date>2019-06-24</date><deptcode>MACS</deptcode><abstract>Deep convolutional auto-encoder (DCAE) allows to obtain useful features via its internal layer and provide an abstracted latent representation, which has been exploited for clustering analysis. DCAE allows a deep clustering method to extract similar patterns in lowerdimensional representation and find idealistic representative centers for distributed data. In this paper, we present a deep clustering model carried out in presence of varying degrees of supervision. We propose a new version of DCAE to include a supervision component. It introduces a mechanism to inject various levels of supervision into the learning process. This mechanism helps to effectively reconcile extracted latent representations and provided supervising knowledge in order to produce the best discriminative attributes. The key idea of our approach is distinguishing the discriminatory power of numerous structures, through varying degrees of supervision, when searching for a compact structure to form robust clusters. We evaluate our model on MNIST, USPS, MNIST fashion, SVHN datasets and show clustering accuracy on different supervisory levels.</abstract><type>Conference Paper/Proceeding/Abstract</type><journal>Computer Analysis of Images and Patterns</journal><volume>11678</volume><paginationStart>224</paginationStart><paginationEnd>233</paginationEnd><publisher/><isbnPrint>978-3-030-29887-6</isbnPrint><isbnElectronic>978-3-030-29888-3</isbnElectronic><issnPrint>0302-9743</issnPrint><issnElectronic>1611-3349</issnElectronic><keywords/><publishedDay>3</publishedDay><publishedMonth>9</publishedMonth><publishedYear>2019</publishedYear><publishedDate>2019-09-03</publishedDate><doi>10.1007/978-3-030-29888-3_18</doi><url/><notes/><college>COLLEGE NANME</college><department>Mathematics and Computer Science School</department><CollegeCode>COLLEGE CODE</CollegeCode><DepartmentCode>MACS</DepartmentCode><institution>Swansea University</institution><apcterm/><lastEdited>2020-08-18T12:40:15.7748643</lastEdited><Created>2019-06-24T11:17:06.9079329</Created><path><level id="1">Faculty of Science and Engineering</level><level id="2">School of Mathematics and Computer Science - Computer Science</level></path><authors><author><firstname>A.</firstname><surname>Alqahtani</surname><order>1</order></author><author><firstname>Xianghua</firstname><surname>Xie</surname><orcid>0000-0002-2701-8660</orcid><order>2</order></author><author><firstname>J.</firstname><surname>Deng</surname><order>3</order></author><author><firstname>Mark</firstname><surname>Jones</surname><orcid>0000-0001-8991-1190</orcid><order>4</order></author></authors><documents><document><filename>0050907-24062019111907.pdf</filename><originalFilename>CAIP-186.pdf</originalFilename><uploaded>2019-06-24T11:19:07.7100000</uploaded><type>Output</type><contentLength>1068531</contentLength><contentType>application/pdf</contentType><version>Accepted Manuscript</version><cronfaStatus>true</cronfaStatus><embargoDate>2020-08-22T00:00:00.0000000</embargoDate><copyrightCorrect>true</copyrightCorrect><language>eng</language></document></documents><OutputDurs/></rfc1807>
spelling 2020-08-18T12:40:15.7748643 v2 50907 2019-06-24 Learning Discriminatory Deep Clustering Models b334d40963c7a2f435f06d2c26c74e11 0000-0002-2701-8660 Xianghua Xie Xianghua Xie true false 2e1030b6e14fc9debd5d5ae7cc335562 0000-0001-8991-1190 Mark Jones Mark Jones true false 2019-06-24 MACS Deep convolutional auto-encoder (DCAE) allows to obtain useful features via its internal layer and provide an abstracted latent representation, which has been exploited for clustering analysis. DCAE allows a deep clustering method to extract similar patterns in lowerdimensional representation and find idealistic representative centers for distributed data. In this paper, we present a deep clustering model carried out in presence of varying degrees of supervision. We propose a new version of DCAE to include a supervision component. It introduces a mechanism to inject various levels of supervision into the learning process. This mechanism helps to effectively reconcile extracted latent representations and provided supervising knowledge in order to produce the best discriminative attributes. The key idea of our approach is distinguishing the discriminatory power of numerous structures, through varying degrees of supervision, when searching for a compact structure to form robust clusters. We evaluate our model on MNIST, USPS, MNIST fashion, SVHN datasets and show clustering accuracy on different supervisory levels. Conference Paper/Proceeding/Abstract Computer Analysis of Images and Patterns 11678 224 233 978-3-030-29887-6 978-3-030-29888-3 0302-9743 1611-3349 3 9 2019 2019-09-03 10.1007/978-3-030-29888-3_18 COLLEGE NANME Mathematics and Computer Science School COLLEGE CODE MACS Swansea University 2020-08-18T12:40:15.7748643 2019-06-24T11:17:06.9079329 Faculty of Science and Engineering School of Mathematics and Computer Science - Computer Science A. Alqahtani 1 Xianghua Xie 0000-0002-2701-8660 2 J. Deng 3 Mark Jones 0000-0001-8991-1190 4 0050907-24062019111907.pdf CAIP-186.pdf 2019-06-24T11:19:07.7100000 Output 1068531 application/pdf Accepted Manuscript true 2020-08-22T00:00:00.0000000 true eng
title Learning Discriminatory Deep Clustering Models
spellingShingle Learning Discriminatory Deep Clustering Models
Xianghua Xie
Mark Jones
title_short Learning Discriminatory Deep Clustering Models
title_full Learning Discriminatory Deep Clustering Models
title_fullStr Learning Discriminatory Deep Clustering Models
title_full_unstemmed Learning Discriminatory Deep Clustering Models
title_sort Learning Discriminatory Deep Clustering Models
author_id_str_mv b334d40963c7a2f435f06d2c26c74e11
2e1030b6e14fc9debd5d5ae7cc335562
author_id_fullname_str_mv b334d40963c7a2f435f06d2c26c74e11_***_Xianghua Xie
2e1030b6e14fc9debd5d5ae7cc335562_***_Mark Jones
author Xianghua Xie
Mark Jones
author2 A. Alqahtani
Xianghua Xie
J. Deng
Mark Jones
format Conference Paper/Proceeding/Abstract
container_title Computer Analysis of Images and Patterns
container_volume 11678
container_start_page 224
publishDate 2019
institution Swansea University
isbn 978-3-030-29887-6
978-3-030-29888-3
issn 0302-9743
1611-3349
doi_str_mv 10.1007/978-3-030-29888-3_18
college_str Faculty of Science and Engineering
hierarchytype
hierarchy_top_id facultyofscienceandengineering
hierarchy_top_title Faculty of Science and Engineering
hierarchy_parent_id facultyofscienceandengineering
hierarchy_parent_title Faculty of Science and Engineering
department_str School of Mathematics and Computer Science - Computer Science{{{_:::_}}}Faculty of Science and Engineering{{{_:::_}}}School of Mathematics and Computer Science - Computer Science
document_store_str 1
active_str 0
description Deep convolutional auto-encoder (DCAE) allows to obtain useful features via its internal layer and provide an abstracted latent representation, which has been exploited for clustering analysis. DCAE allows a deep clustering method to extract similar patterns in lowerdimensional representation and find idealistic representative centers for distributed data. In this paper, we present a deep clustering model carried out in presence of varying degrees of supervision. We propose a new version of DCAE to include a supervision component. It introduces a mechanism to inject various levels of supervision into the learning process. This mechanism helps to effectively reconcile extracted latent representations and provided supervising knowledge in order to produce the best discriminative attributes. The key idea of our approach is distinguishing the discriminatory power of numerous structures, through varying degrees of supervision, when searching for a compact structure to form robust clusters. We evaluate our model on MNIST, USPS, MNIST fashion, SVHN datasets and show clustering accuracy on different supervisory levels.
published_date 2019-09-03T19:45:23Z
_version_ 1821345393815846912
score 11.04748

Similar Items